This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 90/8/866    most recent 01.RES.0000016837.26733.BEv1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fontana, J. Articles by Sessa, W. C. Search for Related Content PubMed PubMed Citation Articles by Fontana, J. Articles by Sessa, W. C. Related Collections Signal transduction Gene therapy Endothelium/vascular type/nitric oxide Cell signalling/signal transduction Genetically altered mice

(Circulation Research. 2002;90:866.)
© 2002 American Heart Association, Inc.

Molecular Medicine

Domain Mapping Studies Reveal That the M Domain of hsp90 Serves as a Molecular Scaffold to Regulate Akt-Dependent Phosphorylation of Endothelial Nitric Oxide Synthase and NO Release

Jason Fontana^*, David Fulton^*, Yan Chen, Todd A. Fairchild, Timothy J. McCabe, Naoya Fujita, Takashi Tsuruo, William C. Sessa

From the Department of Pharmacology and Molecular Cardiobiology Program (J.F., D.F., Y.C., T.A.F., T.J.M., W.C.S.), Boyer Center for Molecular Medicine, Yale University School of Medicine, New Haven, Conn; and the Institute of Molecular and Cellular Biosciences (N.F., T.T.), Bunkyo-ku, Tokyo, Japan.

Correspondence to William C. Sessa, PhD, Dept of Pharmacology and Molecular Cardiobiology Program, Boyer Center for Molecular Medicine, Yale University School of Medicine, 295 Congress Ave, New Haven, CT 06536. E-mail william.sessa{at}yale.edu

	Abstract

Top Abstract Introduction Materials and Methods Results and Discussion References

 
Protein-protein interactions with the molecular chaperone hsp90 and phosphorylation on serine 1179 by the protein kinase Akt leads to activation of endothelial nitric oxide synthase. However, the interplay between these protein-protein interactions remains to be established. In the present study, we show that vascular endothelial growth factor stimulates the coordinated association of hsp90, Akt, and resultant phosphorylation of eNOS. Characterization of the domains of hsp90 required to bind eNOS, using yeast 2-hybrid, cell-based coprecipitation experiments, and GST-fusion proteins, revealed that the M region of hsp90 interacts with the amino terminus of eNOS and Akt. The addition of purified hsp90 to in vitro kinase assays facilitates Akt-driven phosphorylation of recombinant eNOS protein, but not a short peptide encoding the Akt phosphorylation site, suggesting that hsp90 may function as a scaffold for eNOS and Akt. In vivo, coexpression of adenoviral or the cDNA for hsp90 with eNOS promotes nitric oxide release; an effect eliminated using a catalytically functional phosphorylation mutant of eNOS. These results demonstrate that stimulation of endothelial cells with vascular endothelial growth factor recruits eNOS and Akt to an adjacent region on the same domain of hsp90, thereby facilitating eNOS phosphorylation and enzyme activation.

Key Words: nitric oxide • signaling • scaffold • hsp90 • Akt

	Introduction

Top Abstract Introduction Materials and Methods Results and Discussion References

 
Endothelial nitric oxide synthase (eNOS) continually produces low levels of nitric oxide (NO) to regulate several aspects of cardiovascular homeostasis. In endothelial cells and cells transfected with the eNOS cDNA, eNOS behaves as a peripheral membrane protein that is regulated by the allosteric activator, calmodulin (CaM). In vitro, the addition of CaM to recombinant eNOS markedly accelerates NOS catalytic function and NO synthesis.¹ However, in vivo, additional regulatory mechanisms other than CaM participate in eNOS activation/inactivation. This concept is supported by studies demonstrating that mislocalization of eNOS secondary to mutations that block its membrane association do not influence its catalytic function or calcium dependency in vitro; however, agonist-stimulated NO release from cells is markedly diminished.^2–4 These studies imply that spatial and temporal regulation of membrane associated eNOS function must involve other protein-protein or protein-lipid interactions that impact on its activation state.

In the past several years, many protein partners that interact with eNOS have been described, including caveolins-1 and -3,^5,6 heat shock protein 90 (hsp90),⁷ dynamin-2,⁸ G protein–coupled receptors,⁹ and certain kinases including Akt and mitogen-activated protein kinase family members.^10,11 All these proteins have been show to interact with eNOS in conventional in vitro assays, including coprecipitations, affinity chromatography, and yeast 2-hybrid analysis. In vivo, there is compelling evidence supporting the importance of caveolin-1, hsp90, and Akt in regulating NO release because overexpression, inhibition, or dominant-negative strategies to each protein partner blocks NO release and endothelium-dependent vasodilation in intact blood vessels.^7,12–16 With this in mind, a reasonable working model is that caveolin-1 binding to eNOS renders it less active. On stimulation of cells with various agonists, CaM and hsp90 are recruited, the caveolin-1 inhibitory clamp is displaced, and phosphorylation by Akt ensues to modulate eNOS catalysis and NO release.¹⁷ However, the precise interrelationships and mechanisms of how these protein regulators modulate eNOS are not well understood.

Because hsp90 and Akt are both involved with eNOS activation, the purposes of this study are to (1) evaluate the interaction of hsp90 with Akt and eNOS, (2) map the binding domains of Akt and eNOS on hsp90 using multiple methodologies, and (3) determine the functional significance of this complex in vitro and in vivo.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results and Discussion References

 
Cell Culture/Western Blotting
Bovine aortic endothelial cells (BAECs, passage 4 or less) were grown in Dulbecco’s modified Eagle’s medium containing penicillin (100 U/mL), streptomycin (100 mg/mL), and 10% (v/v) fetal calf serum (complete Dulbecco’s modified Eagle’s medium). BAECs were serum starved for 12 hours overnight, stimulated with vascular endothelial growth factor (VEGF; 50ng/mL) for the times indicated, and protein immunoprecipitated, as described previously.¹³

In Vitro Kinase Assays
For in vitro phosphorylation studies, we incubated recombinant eNOS (1 µg) purified from Escherichia Coli with activated recombinant Akt from baculovirus (12 ng, Upstate Biotechnologies) in the presence or absence of purified bovine brain hsp90 (2 µg, Stressgen) or denatured hsp90 (boiled). Proteins were preincubated on ice for 15 minutes and kinase assays were performed in a buffer containing HEPES (20 mmol/L, pH 7.4), MnCl₂ (10 mmol/L), MgCl₂ (10 mmol/L) with ATP (50 µmol/L) and DTT (1 mmol/L) for the times indicated. For experiments using eNOS peptides as substrates for Akt, WT peptide (20 µg, RIRTQSFSLQERHLRGAVPWA) and SA peptide (RIRTQAFSLQERHLRGAVPWA) were preincubated with hsp90 (as above), and kinase assays were performed in the presence of ³²P -ATP (1 µL, specific activity 3000 Ci/mmol) and ATP (50 µmol/L) for 5 minutes. Reactions were then spotted onto phosphocellulose filters and the amount of phosphate incorporated was measured by Cerenkov counting.

Yeast Two-Hybrid Analysis of Interacting Domains
pAS2-1 (DNA binding domain hybrid cloning vector), pACT2 (activation domain hybrid cloning vector), and Saccharomyces cerevisiae Y190 were purchased from Clontech. cDNA constructs encoding hybrids of bovine eNOS and the GAL4 DNA binding domain and Hsp90ß and GAL4 activation domain were created by subcloning into the expression vectors pAS2-1 and pACT2, respectively. S. cerevisiae strain Y190 was cotransformed by the lithium acetate method with various pair-wise combinations of pAS2-1 and pACT2 fusion plasmids. Transformants were grown on plates lacking histidine (in the presence of 2 mmol/L 3-aminotriazole), tryptophan and leucine to select for the presence of the pAS2-1 and pACT2 derivatives, respectively. After 3 to 4 days, transformants were assayed for growth and ß-galactosidase activity by the colony lift filter assay. All DNA-binding domain and activation domain fusion constructs were confirmed not to activate reporter gene transcription by themselves, and all activation domain fusion constructs were confirmed not to activate transcription when combined with the unrelated pLAM 5'-1 (human lamin C in pAS2-1).

Mapping of the hsp90/eNOS Interaction in COS-7 Cells
The Flag-tagged hsp90 deletion constructs and theV5-tagged eNOS constructs were generated by using standard cloning methods. COS-7 cells were plated in 60-mm dishes and transfected with eNOS (2 to 3 µg), HA-hsp90 (2 to 3 µg), Flag-hsp90 constructs (3 µg), or V5-eNOS deletion constructs (2 µg). To balance all transfections, the expression vector for ß-galactosidase cDNA was used. Twenty-four hours after transfection, epitope-tagged constructs were immunoprecipitated as described above, using 2 µg of anti-flag monoclonal antibody (M2) (Sigma) or anti-HA rat monoclonal antibody (Boehringer Mannheim), respectively.

Purification of GST-hsp90 Fusion Protein and In Vitro Interactions
The hsp90 fusion protein expression was performed as described previously.¹⁸ The interaction of GST-hsp90 fusion proteins [N (9–236), M (272–617), or C domains (629–732)] with recombinant eNOS and Akt and proteins from BAEC cell lysates was performed as described in the expanded Materials and Methods section that can be found in the online data supplement available at http://www.circresaha.org.

Cos-7 Cell Transfections and NO Release
The epitope-tagged cDNAs for bovine eNOS, S1179A eNOS, and HA-tagged hsp90 constructs¹⁹ were generated by standard cloning methods. COS-7 cells were plated in 60-mm dishes and transfected with eNOS (2 to 3 µg), HA-hsp90 (2 to 3 µg), and Flk (3 µg). To balance all transfections, the expression vector for ß-galactosidase cDNA was used. Twenty-four hours after transfection, the expression of appropriate proteins was confirmed. In some experiments, COS-7 cells were serum starved for 12 to 16 hours and stimulated with VEGF (50ng/mL) for the times indicated. Lysates were prepared for Western blotting as discussed above and NO release was measured as described in the expanded Materials and Methods section.

Production of Adenoviral hsp90
Replication-deficient adenoviruses expressing HA-tagged hsp90 or ß-galactosidase under the control of the cytomegalovirus (CMV) promoter were generated using the pAdTrack-CMV vector and the AdEasy System. The viruses were amplified in HEK293 cells, purified using CsCl, and titered using a cytopathic effect (CPE). Infection with 50 MOI of viruses resulted in close to 100% of the cells expressing the gene of interest with no signs of toxicity.

	Results and Discussion

Top Abstract Introduction Materials and Methods Results and Discussion References

 
VEGF has been shown to stimulate the recruitment of hsp90⁷ and promote Akt-dependent phosphorylation of eNOS leading to NO production.^11,13 Therefore, we initially investigated the time course of VEGF-stimulated signaling to eNOS. Treatment of bovine aortic endothelial cells with VEGF (50 ng/mL) stimulated the time-dependent phosphorylation of eNOS on serine 1179 and Akt phosphorylation on serine 473 with maximal activation occurring at 3 minutes (Figure 1A). Accordingly, VEGF rapidly stimulated the recruitment of hsp90 to eNOS in a similar time frame (Figure 1B), implying that these contemporaneous events may occur in a multiprotein complex. Because previous studies have shown that Akt can physically interact with either eNOS^10,11 or hsp90¹⁹ independently, we examined whether VEGF can promote hsp90-dependent recruitment of activated Akt and phosphospecific (P-) eNOS. As seen in Figure 1C, immunoprecipitation of hsp90 in nonstimulated cells results in little recovery of P-eNOS and P-Akt. However, on VEGF challenge, immunoprecipitation of hsp90 results in more activated eNOS (phospho-1179) and Akt (phospho-473) in the immunocomplex overtime consistent with the time courses in Figures 1A and 1B. These data suggest the stimulus-dependent formation of a multiprotein complex containing hsp90, Akt, and eNOS.

View larger version (33K): [in this window] [in a new window]   Figure 1. VEGF simultaneously activates Akt-dependent eNOS phosphorylation and the association of hsp90, Akt, and eNOS. A, Serum starved BAECs were stimulated with VEGF (50ng/mL) for different time points and the level of eNOS and Akt phosphorylation assessed using phosphospecific antibodies (P-eNOS and P-Akt). B, VEGF-stimulated association of hsp90 to eNOS was examined by immunoprecipitation of eNOS at various time points. C, Assembly of activated eNOS and Akt was determined after VEGF treatment and immunoisolation of hsp90. hsp90 immunoprecipitates were Western blotted for P-eNOS and P-Akt. D-units (densitometric units) reflect relative density of phospho-protein to total eNOS or Akt in A, hsp90 to eNOS in B, and P-eNOS and P-Akt to hsp90 in C. Similar results were obtained in 2 to 4 additional experiments.

The specific domains of hsp90 involved in protein binding are beginning to be elucidated. In general, adenine nucleotides and geldanamycin bind in the ATP binding pocket of the amino terminus; calmodulin, calponin, the glucorticoid receptor, and most recently, Akt, bind near the charged region in the middle domain (M domain)¹⁹; and proteins containing tetricopeptide repeats (such as immunophilin and PP5) bind to the carboxy terminus.²⁰ Nonnative proteins have been shown to bind either to the N or C terminus, but holo-hsp 90 is more effective than either domain in protein refolding.²¹ In order to study the interactions of eNOS with hsp90, we used 3 distinct approaches: yeast 2-hybrid analysis, coprecipitation experiments, and in vitro protein-protein interaction. cDNA fragments encoding different domains of hsp90ß (25–232, 232–442, 442–600, and 600–732) were cloned in a vector containing the binding domain (BD) of the GAL4 transcription factor, and fragments of eNOS (2–403, 304–709, and 1006–1205) were cloned into the activation domain (AD) vector and were tested in a 2-hybrid screen. As seen in the Table, using both growth in synthetic dropout medium and ß-galactosidase activity as index of a successful interaction, hsp90 domains did not interact with eNOS 304–709 or 1006–1205. hsp90 domains 25–232, 232–442, and 600–732 weakly interacted with the amino terminal fragment of eNOS (2–403). However, the region of the M domain (442–600) strongly interacts with eNOS (2–403). Because eNOS (340–709) did not interact with any hsp90 construct, we conclude that the primary binding domain on eNOS for hsp90 is between amino acids 2–304, and the primary binding domain in hsp90 for eNOS are amino acids 442–600. All other constructs tested were essentially negative for growth and ß-galactosidase activity as indicated in the Table.

View this table: [in this window] [in a new window]   Table 1. eNOS/hsp90 Interactions in Yeast Two-Hybrid Analysis

Next, we used a series of deletion mutants of either hsp90 or eNOS to detect which domains of interaction were required in intact cells. COS cells were transfected with the cDNAs for wild-type eNOS and flag-tagged hsp90 (Figure 2A) or V5-tagged eNOS and HA-tagged hsp90 and the interaction assessed in immunoprecipitation experiments. In both panels, the transfected cDNAs were equally immunoprecipitated and expressed (see middle and bottom panels in Figures 2A and 2B). As seen in Figure 2A, cotransfection of different flag-tagged, hsp90 deletion mutants with full-length eNOS resulted in eNOS coprecipitating with regions 1–456 and 1–530, but not 1–309 or 530–724 of hsp90. This suggests that regions of hsp90 between amino acids 309 and 530 are important for interacting with eNOS. In Figure 2B, the domains of eNOS that interact with hsp90 were determined similarly. Cotransfection of different V5-tagged, eNOS deletion mutants with full-length HA-tagged hsp90 resulted in hsp90 coprecipitation with regions 1–400 and 1–600, but not 1–300 of eNOS. Thus, regions between amino acids 300 and 400 of eNOS are sufficient to interact with hsp90 in cells.

View larger version (37K): [in this window] [in a new window]   Figure 2. In vivo mapping of the eNOS-hsp90 interacting domains. A, COS cells were cotransfected with the cDNAs encoding flag-tagged hsp90 (amino acids 1–309, 1–456, 1–530, and 530–724) and wild-type eNOS. The flag-epitope was immunoprecipitated and Western blotted for eNOS complexed (top), hsp90 mutant immunoprecipitated (middle), and expression of eNOS in the cell lysate (bottom). B, COS cells were cotransfected with the cDNAs encoding V5-tagged eNOS (amino acids 1–100, 1–200, 1–300, 1–400, and 1–600) and HA-tagged hsp90. The HA-epitope was immunoprecipitated and Western blotted for eNOS complexed (top), hsp90 immunoprecipitated (middle), and expression of eNOS constructs in the cell lysate (bottom). IgGL refers to the light chain of the antibody used for immunoprecipitation.

To test the interactions using an additional approach, we used recombinant eNOS or endothelial cell lysate as a source of eNOS and GST fusion proteins spanning the N (9–236), M (272–617) and C domains (629–732) of hsp90. These regions of hsp90 are greater than 95% identical to corresponding amino acids 9–231, 266–610, and 622–724 of hsp90ß. As seen is Figure 3A, incubation of eNOS with GST alone, N, M, or C domains of hsp90 in the presence 150 mmol/L NaCl revealed a preferential interaction between eNOS and the M domain of hsp90 with low-level binding to the N and C domains (Figure 3A, left panel). Similarly, in BAEC lysates, eNOS interacted strongly with the M domain again, with low-level binding to the N and C domains (right domain). In the absence of NaCl in the binding and wash buffers, eNOS interacted more strongly with all domains, but not GST alone (data not shown). Collectively, our data from the 2-hybrid analysis and GST binding assays demonstrate that the amino acids 442–600 in hsp90ß or 449–607 in hsp90 in the M domain are responsible for binding the amino terminus of eNOS.

View larger version (35K): [in this window] [in a new window]   Figure 3. eNOS and Akt both bind to the M domain of hsp90. A, Incubation of recombinant eNOS (left) or eNOS from endothelial cell lysates (right) with GST, GST-N, -M, and C domains of hsp90 results in preferential interaction of eNOS with the M domain of hsp90 (n=3 individual experiments). B, Recombinant Akt (left) or Akt from endothelial cell lysates preferentially interacts with the M domain of hsp90 (n=3 experiments). The low-level interaction with the C domain was consistent throughout the studies. In C, GST, GST-N, -M, and -C domains were preincubated with eNOS and washed. Activated Akt was then added to the in vitro kinase reaction and P-eNOS determined by Western blotting (n=3).

Recently, it has been shown that Akt can be immunoisolated in a complex with hsp90 from cell lysates via a direct interaction between residues 229–309 of Akt and residues 327–340 of the M domain of hsp90.¹⁹ In cells, the binding of hsp90 to Akt prevents dephosphorylation of threonine 308 by reducing PP2A-mediated dephosphorylation. To verify the interaction between hsp90 and Akt in our system, we incubated recombinant Akt or endothelial cell lysates with GST-N, -M, and -C domains of hsp90. As seen in Figure 3B, recombinant Akt (left panel) or Akt from cell lysates (right panel) does not significantly interact with GST alone, but interacts with the M domain of hsp90 and binds to a lesser extent to the C domain. The low-level binding to the C domain may represent an additional site of interaction to stabilize Akt to hsp90. The binding of Akt to residues 327–340 and eNOS to residues 442–600 of hsp90ß of the M domain suggests that hsp90 can serve as a scaffold for the kinase and it substrate. To test if Akt can phosphorylate eNOS bound to the M domain of hsp90, recombinant Akt and eNOS were preincubated with either GST alone or M domain of hsp90 and an in vitro kinase reaction performed. As seen in Figure 3C, Akt can phosphorylate eNOS bound to the M domain. These data define (1) the binding motif of hsp90 on eNOS, (2) Akt and eNOS interacting with a common domain of hsp90 in a nonoverlapping manner, and (3) the ability of Akt to phosphorylate eNOS bound to the M domain of hsp90.

The above data documenting the bipartite binding of Akt and eNOS to a specific domain of hsp90 suggests that perhaps binding to hsp90 may influence the ability of eNOS to serve as an Akt substrate or that docking to hsp90 may influence Akt activity. To see if eNOS was a better Akt substrate in the presence of hsp90, in vitro kinase reactions were performed with recombinant eNOS and Akt, in the absence or presence of hsp90 in the reaction mix, and the extent of P-eNOS examined. As seen in Figure 4A, the addition of hsp90 to the in vitro kinase reaction increased the detection of P-eNOS after only 1 minute of incubation and phosphorylation increased further at 3 minutes compared with reactions lacking hsp90 (as quantified in Figure 4B). However, at later time points, the level of P-eNOS was slightly reduced in the presence of hsp90. Furthermore, thermal denaturation of hsp90 abrogated the ability of hsp90 to increase Akt-dependent phosphorylation of eNOS (Figure 4C) but did not completely eliminate the ability of Akt to phosphorylate eNOS. Next, we examined if the docking of Akt to hsp90 influenced its kinase activity by examining the ability of Akt to phosphorylate a NOS peptide (¹¹⁷⁴IRTQS*FSLQERHLRGAVPWA¹¹⁹⁴) containing the Akt phosphorylation site (serine 1179 with asterisk). We rationalized that, because the Akt phosphorylation site is at the extreme carboxy tail of eNOS, the peptide would serve as a substrate and would not interact with hsp90. As seen in Figure 3D, the addition of Akt to the wild-type eNOS peptide increased the amount of ³²P incorporated into the peptide (lane 1) compared with a mutant peptide where S1179 was mutated to an alanine residue (S1179A) as previously described.¹³ However, the addition of hsp90 to the reaction did not increase the amount of ³²P incorporated. This suggests that the docking of eNOS to hsp90 induces a conformation favorable for phosphorylation by Akt, but hsp90, per se, does not increase the activity of Akt toward eNOS as a substrate.

View larger version (13K): [in this window] [in a new window]   Figure 4. The interaction of eNOS and Akt with hsp90 increases phosphorylation on serine 1179. A, Recombinant eNOS was incubated with activated Akt, in the absence or presence of hsp90, and an in vitro kinase reaction performed over 10 minutes. Aliquots of the reactions were blotted for phosphospecific eNOS (P-eNOS), total eNOS, phosphospecific Akt (P-Akt), and hsp90 (left). B, Quantitation of P-eNOS/total eNOS levels over time. Similar experiments were repeated 3 additional times. Data in B is a graphic representation of the densitometric analysis of data in A. C, Conditions were identical to A, except hsp90 was heat denatured before adding it to the in vitro kinase reaction. D, Wild-type (WT) eNOS peptide or eNOS S1179A peptide was used as a substrate for in vitro kinase reactions in the absence or presence of hsp90. The presence of hsp90 did not increase the level of eNOS peptide phosphorylation, suggesting that docking of the entire protein was important for its ability to act as a substrate for Akt. Data are mean±SEM; n=6.

To examine if overexpression of hsp90 would influence eNOS phosphorylation on serine 1179 and regulate NO release, COS cells were cotransfected with the cDNAs for VEGF receptor Flk-1, HA-tagged hsp90ß, and either wild-type eNOS (in Figure 5A) or eNOS S1179A (in Figure 5B). As seen in Figure 5A, the presence of hsp90, but not ß-galactosidase, increased both basal and VEGF-stimulated eNOS phosphorylation on serine 1179. Next, we examined VEGF-stimulated NO production in cells expressing either wild-type eNOS or eNOS S1179A. VEGF-stimulated NO production was enhanced by hsp90 cotransfection with wild-type eNOS. In contrast, mutation of serine 1179 to alanine abrogated the ability of hsp90 to influence NO release.

View larger version (20K): [in this window] [in a new window]   Figure 5. hsp90 promotes eNOS phosphorylation and VEGF-stimulated NO production in a reconstituted system. A, COS cells were transfected with cDNAs encoding Flk-1, wild-type eNOS (WT), and hsp90ß or the control plasmid, ß-galactosidase, and eNOS phosphorylation on S1179 examined by Western blotting. B, Flk-1, wild-type eNOS (WT) or S1179A eNOS, and hsp90ß or the control plasmid, ß-galactosidase was transfected for NO release. The levels of Flk, eNOS, and HA-hsp90 were identical throughout. Cells were then stimulated with PBS (C for control) or VEGF (50 ng/mL) for 30 minutes and VEGF-stimulated NO2- quantified by NO specific chemiluminescence. NO levels from ß-gal transfected COS cells were subtracted as background. Data are mean±SEM with n=6; P<0.05. Similar data were found in 3 additional experiments.

Finally, to see if hsp90 regulated Akt and eNOS in primary cultures of BAECs, we developed an adenovirus expressing HA-tagged hsp90 (Ad hsp90) and examined basal and VEGF-stimulated eNOS and Akt phosphorylation and NO release. As seen in Figure 6A, in ß-galactosidase–infected cells, eNOS was basally phosphorylated on serines 1179 and 497, with low-level Akt phosphorylation and VEGF-enhanced eNOS phosphorylation on S1179 and Akt phosphorylation. Interestingly, transduction of BAECs with Ad hsp90 increased basal phosphorylation of eNOS (on S1179 and T497) and Akt, and the addition of VEGF further enhanced phosphorylation of eNOS on S1179 and Akt. We then quantified basal and VEGF-stimulated NO release from transduced BAECs. Transduction of BAECs with Ad hsp90 increased basal NO release over 16 hours compared with cells infected with Ad ß-galactosidase. Moreover, VEGF-stimulated NO production was also enhanced by the presence of hsp90. Collectively, in both a reconstituted COS cell system and cultures of BAECs, our data support the idea that hsp90 enhances basal and VEGF-stimulated Akt and eNOS phosphorylation on serine 1179 and, subsequently, NO release and underscores the importance of phosphorylation of serine 1179 for stimulated NO release.

View larger version (30K): [in this window] [in a new window]   Figure 6. Adenoviral expression of hsp90 promotes eNOS phosphorylation and NO release from endothelial cells. A, BAECs were infected with adenoviral ß-galactosidase (Ad ß-gal, left) or Ad hsp90 (right) and lysates analyzed by Western blotting with appropriate antibodies. B, NO release accumulating over 14 hours (left) or after stimulation with VEGF (50 ng/mL for 30 minutes) was determined by NO-specific chemiluminescence. Data are mean±SEM with n=6; P<0.05.

This study integrates 2 pathways previously implicated in the activation of eNOS, namely hsp90 and Akt. We provide evidence that hsp90 can serve as a scaffold to promote a productive interaction between Akt and the substrate eNOS due to the proximity of these proteins once complexed with hsp90. Previously, we and others have shown that hsp90 is recruited to eNOS in a time frame consistent with NO release and that hsp90 overexpression increases NOS activity.^7,22,23 Very recently, Brouet et al¹⁷ have elegantly shown that hsp90 is necessary for Akt-dependent eNOS phosphorylation in coprecipitation studies. However, the mechanism and role of hsp90 as scaffold were not examined. Our study confirms and significantly extends these findings by showing the precise modular domains of hsp90 that accommodate the docking of Akt and eNOS, respectively, and shows that this proximity model permits eNOS to be phosphorylated more efficiently. In addition, we show that adenoviral expression of hsp90 can increase basal and stimulated Akt phosphorylation, presumably by blocking its dephosphorylation as described,¹⁹ promoting basal and stimulated eNOS phosphorylation on serine 1179, and enhancing NO release. Surprisingly, eNOS phosphorylation on threonine 497 is increased by Ad hsp90, suggesting that hsp90 may promote phosphorylation or prevent dephosphorylation of this site. Regardless of the mechanism, it does not appear that changes in threonine 497 phosphorylation/dephosphorylation influences the magnitude of basal and VEGF-stimulated NO release in this study. These results examining threonine 497 phosphorylation/dephosphorylation, in the context of hsp90 and Akt, are similar to the findings of Brouet et al.²⁴ These authors demonstrated that overexpression of hsp90 promotes eNOS phosphorylation and NO-dependent capillary tube formation in endothelial cells. The similarity of our results with Ad hsp90 provide further support for the idea that overexpression of hsp90 within endothelial cells promotes eNOS activation and NO-dependent functions of the endothelium.

In vitro, several mechanisms have been proposed to explain how hsp90 influences NOS function including allosteric modulation of NOS,⁷ coupling of L-arginine utilization to NO synthesis,²³ enhancing CaM affinity for NOS,^25,26 and increasing heme incorporation into NOS.²⁷ Indeed, the ability of hsp90 to participate in protein folding and stabilization reactions as well as signal transduction cascades gives credence to its role as an evolutionary capacitor in yeast.²⁸ Moreover, the identification of hsp90 as a central component in the activation/phosphorylation of eNOS suggests that hsp90 may be involved with the regulation of other Akt substrates involved in cell survival pathways and metabolism.

Thus, based on several studies and the present data, we propose the following model to explain growth factor–stimulated eNOS activation (see Figure 7): (1) VEGF receptor engagement leads to activation of the PLC and PI-3 kinase/Akt pathways; (2) calcium-activated CaM participates with hsp90 in disrupting the effects of caveolin-1 on eNOS; (3) the interaction of hsp90 with Akt and eNOS permits hsp90 to serve as a docking site for Akt-dependent phosphorylation of eNOS; (4) the binding of CaM and phosphorylation of eNOS on serine 1179, effects stabilized by hsp90, leads to enhanced electron flux from the reductase to the oxygenase domain of eNOS; and (5) NO release occurs. The discovery of threonine 495 dephosphorylation as a mechanism to stabilize/enhance the interactions of CaM with eNOS^29–31 most likely occurs in step 4; however, there is little evidence that dephosphorylation occurs in response to VEGF,^17,29,32 but this step is clearly pertinent for other agonists such as bradykinin. Further molecular dissection of this complex model will pave the way for both a structural and functional understanding of how endothelial cells manufacture NO.

View larger version (24K): [in this window] [in a new window]   Figure 7. Proposed model for eNOS regulation by posttranslational modifications. In quiescent endothelial cells (left), eNOS can be found in a complex with caveolin-1, a negative regulator, and low levels of hsp90, which may promote biosynthesis and conformational congruency of the enzyme. On stimulation of cells with VEGF (right), simultaneous increases in calcium and activation of PI-3 kinase results in (1) contemporaneous removal of the caveolin inhibitory clamp with CaM and hsp90 recruitment and (2) hsp90 recruitment facilitating the docking of Akt and phosphorylation of eNOS. Thus, phosphorylation on S1179 in the presence of bound CaM are necessary for VEGF-stimulated NO release. The functional consequence of hsp90 on T497 phosphorylation and the role of this residue in activation of eNOS by VEGF is not known.

	Acknowledgments

 
This work is supported by grants from the National Institute of Health (RO1 HL57665, HL61371, HL64793 to W.C.S. and T32HL10183 to D.F.) and a Grant-in-Aid from the American Heart Association (National Grant to W.C.S.). T.A.F. is a Howard Hughes Medical Institute Medical Student Research Training Fellow. We would like to thank Dr Ullrich Hartl for the GST hsp90 constructs. This study was performed while W.C. Sessa was an Established Investigator of the American Heart Association.

	Footnotes

 
*Both authors contributed equally to this work.

Received November 28, 2001; revision received March 14, 2002; accepted March 15, 2002.

	References

Top Abstract Introduction Materials and Methods Results and Discussion References

 

1. Pollock JS, Forstermann U, Mitchell JA, Warner TD, Schmidt HH, Nakane M, Murad F. Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proc Natl Acad Sci U S A. 1991; 88: 10480–10484.
2. Sessa WC, García-Cardeña G, Liu J, Keh A, Pollock JS, Bradley J, Thiru S, Braverman IM, Desai KM. The Golgi association of endothelial nitric oxide synthase is necessary for the efficient synthesis of nitric oxide. J Biol Chem. 1995; 270: 17641–17644.
3. Sakoda T, Hirata K, Kuroda R, Miki N, Suematsu M, Kawashima S, Yokoyama M. Myristoylation of endothelial cell nitric oxide synthase is important for extracellular release of nitric oxide. Mol Cell Biochem. 1995; 152: 143–148.
4. Liu J, Garcia-Cardena G, Sessa WC. Palmitoylation of endothelial nitric oxide synthase is necessary for optimal stimulated release of nitric oxide: implications for caveolae localization. Biochemistry. 1996; 35: 13277–13281.
5. García-Cardeña G, Fan R, Stern DF, Liu J, Sessa WC. Endothelial nitric oxide is regulated by tyrosine phosphorylation and interacts with caveolin-1. J Biol Chem. 1996; 271: 27237–27240.
6. Feron O, Belhassen L, Kobzik L, Smith TW, Kelly RA, Michel T. Endothelial nitric oxide synthase targeting to caveolae: specific interactions with caveolin isoforms in cardiac myocytes and endothelial cells. J Biol Chem. 1996; 271: 22810–22814.
7. Garcia-Cardena G, Fan R, Shah V, Sorrentino R, Cirino G, Papapetropoulos A, Sessa WC. Dynamic activation of endothelial nitric oxide synthase by Hsp90. Nature. 1998; 392: 821–824.
8. Cao S, Yao J, McCabe TJ, Yao Q, Katusic ZS, Sessa WC, Shah V. Direct interaction between endothelial nitric oxide synthase and dynamin-2: implications for nitric oxide synthase function. J Biol Chem. 2001; 276: 14249–14256.
9. Golser R, Gorren AC, Leber A, Andrew P, Habisch HJ, Werner ER, Schmidt K, Venema RC, Mayer B. Interaction of endothelial and neuronal nitric-oxide synthases with the bradykinin B2 receptor: binding of an inhibitory peptide to the oxygenase domain blocks uncoupled NADPH oxidation. J Biol Chem. 2000; 275: 5291–5296.
10. Bernier SG, Haldar S, Michel T. Bradykinin-regulated interactions of the mitogen-activated protein kinase pathway with the endothelial nitric-oxide synthase. J Biol Chem. 2000; 275: 30707–30715.
11. Michell BJ, Griffiths JE, Mitchelhill KI, Rodriguez-Crespo I, Tiganis T, Bozinovski S, de Montellano PR, Kemp BE, Pearson RB. The Akt kinase signals directly to endothelial nitric oxide synthase. Curr Biol. 1999; 9: 845–848.
12. Garcia-Cardena G, Martasek P, Masters BS, Skidd PM, Couet J, Li S, Lisanti MP, Sessa WC. Dissecting the interaction between nitric oxide synthase (NOS) and caveolin. Functional significance of the nos caveolin binding domain in vivo. J Biol Chem. 1997; 272: 25437–25440.
13. Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF, Papapetropoulos A, Sessa WC. Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature. 1999; 399: 597–601.
14. Luo Z, Fujio Y, Kureishi Y, Rudic RD, Daumerie G, Fulton D, Sessa WC, Walsh K. Acute modulation of endothelial Akt/PKB activity alters nitric oxide–dependent vasomotor activity in vivo. J Clin Invest. 2000; 106: 493–499.
15. Bucci M, Gratton JP, Rudic RD, Acevedo L, Roviezzo F, Cirino G, Sessa WC. In vivo delivery of the caveolin-1 scaffolding domain inhibits nitric oxide synthesis and reduces inflammation. Nat Med. 2000; 6: 1362–1367.
16. Feron O, Dessy C, Opel DJ, Arstall MA, Kelly RA, Michel T. Modulation of the endothelial nitric-oxide synthase-caveolin interaction in cardiac myocytes. Implications for the autonomic regulation of heart rate. J Biol Chem. 1998; 273: 30249–30254.
17. Brouet A, Sonveaux P, Dessy C, Balligand JL, Feron O. Hsp90 ensures the transition from the early Ca²⁺-dependent to the late phosphorylation-dependent activation of the endothelial nitric-oxide synthase in vascular endothelial growth factor-exposed endothelial cells. J Biol Chem. 2001; 276: 32663–32669.
18. Young JC, Schneider C, Hartl FU. In vitro evidence that hsp90 contains two independent chaperone sites. FEBS Lett. 1997; 418: 139–143.
19. Sato S, Fujita N, Tsuruo T. Modulation of Akt kinase activity by binding to Hsp90. Proc Natl Acad Sci U S A. 2000; 97: 10832–10837.
20. Pratt WB. The role of the hsp90-based chaperone system in signal transduction by nuclear receptors and receptors signaling via MAP kinase. Annu Rev Pharmacol Toxicol. 1997; 37: 297–326.
21. Pearl LH, Prodromou C. Structure and in vivo function of Hsp90. Curr Opin Struct Biol. 2000; 10: 46–51.
22. Feron O, Dessy C, Desager JP, Balligand JL. Hydroxy-methylglutaryl-coenzyme: a reductase inhibition promotes endothelial nitric oxide synthase activation through a decrease in caveolin abundance. Circulation. 2001; 103: 113–118.
23. Pritchard KA Jr, Ackerman AW, Gross ER, Stepp DW, Shi Y, Fontana JT, Baker JE, Sessa WC. Heat shock protein 90 mediates the balance of nitric oxide and superoxide anion from endothelial nitric-oxide synthase. J Biol Chem. 2001; 276: 17621–17624.
24. Brouet A, Sonveaux P, Dessy C, Moniotte S, Balligand J-L, Feron O. Hsp90 and caveolin are key targets for the proangiogenic nitric oxide–mediated effects of statins. Circ Res. 2001; 89: 866–873.
25. Song Y, Zweier JL, Xia Y. Heat-shock protein 90 augments neuronal nitric oxide synthase activity by enhancing Ca²⁺/calmodulin binding. Biochem J. 2001; 355: 357–360.
26. Gratton JP, Fontana J, O’Connor DS, Garcia-Cardena G, McCabe TJ, Sessa WC. Reconstitution of an endothelial nitric-oxide synthase (eNOS), hsp90, and caveolin-1 complex in vitro: evidence that hsp90 facilitates calmodulin-stimulated displacement of eNOS from caveolin-1. J Biol Chem. 2000; 275: 22268–22272.
27. Bender AT, Silverstein AM, Demady DR, Kanelakis KC, Noguchi S, Pratt WB, Osawa Y. Neuronal nitric-oxide synthase is regulated by the Hsp90-based chaperone system in vivo. J Biol Chem. 1999; 274: 1472–1478.
28. Rutherford SL, Lindquist S. Hsp90 as a capacitor for morphological evolution. Nature. 1998; 396: 336–342.
29. Michell BJ, Chen Z, Tiganis T, Stapleton D, Katsis F, Power DA, Sim AT, Kemp BE. Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase. J Biol Chem. 2001; 276: 17625–17628.
30. Harris MB, Ju H, Venema VJ, Liang H, Zou R, Michell BJ, Chen ZP, Kemp BE, Venema RC. Reciprocal phosphorylation and regulation of endothelial nitric-oxide synthase in response to bradykinin stimulation. J Biol Chem. 2001; 276: 16587–16591.
31. Fleming I, Fisslthaler B, Dimmeler S, Kemp BE, Busse R. Phosphorylation of Thr⁴⁹⁵ regulates Ca²⁺/calmodulin-dependent endothelial nitric oxide synthase activity. Circ Res. 2001; 88: e68–e75.
32. Fulton D, Fontana J, Sowa G, Gratton JP, Lin M, Li KX, Michell B, Kemp BE, Rodman D, Sessa WC. Localization of endothelial nitric oxide synthase phosphorylated on serine 1179 and nitric oxide in Golgi and plasma membrane defines the existence of two pools of active enzyme. J Biol Chem. 2002; 277: 4277–4284.

This article has been cited by other articles:

				Z. Hu, J. Chen, Q. Wei, and Y. Xia Bidirectional Actions of Hydrogen Peroxide on Endothelial Nitric-oxide Synthase Phosphorylation and Function: CO-COMMITMENT AND INTERPLAY OF Akt AND AMPK J. Biol. Chem., September 12, 2008; 283(37): 25256 - 25263. [Abstract] [Full Text] [PDF]

				L. Belova, D. R. Brickley, B. Ky, S. K. Sharma, and S. D. Conzen Hsp90 Regulates the Phosphorylation and Activity of Serum- and Glucocorticoid-regulated Kinase-1 J. Biol. Chem., July 4, 2008; 283(27): 18821 - 18831. [Abstract] [Full Text] [PDF]

				M. A. Alaamery and C. S. Hoffman Schizosaccharomyces pombe Hsp90/Git10 Is Required for Glucose/cAMP Signaling Genetics, April 1, 2008; 178(4): 1927 - 1936. [Abstract] [Full Text] [PDF]

				K.-J. Yang, S. Shin, L. Piao, E. Shin, Y. Li, K. A. Park, H. S. Byun, M. Won, J. Hong, G. R. Kweon, et al. Regulation of 3-Phosphoinositide-dependent Protein Kinase-1 (PDK1) by Src Involves Tyrosine Phosphorylation of PDK1 and Src Homology 2 Domain Binding J. Biol. Chem., January 18, 2008; 283(3): 1480 - 1491. [Abstract] [Full Text] [PDF]

				R. Q. Miao, J. Fontana, D. Fulton, M. I. Lin, K. D. Harrison, and W. C. Sessa Dominant-Negative Hsp90 Reduces VEGF-Stimulated Nitric Oxide Release and Migration in Endothelial Cells Arterioscler. Thromb. Vasc. Biol., January 1, 2008; 28(1): 105 - 111. [Abstract] [Full Text] [PDF]

				S. Sharma, N. Sud, D. A. Wiseman, A. L. Carter, S. Kumar, Y. Hou, T. Rau, J. Wilham, C. Harmon, P. Oishi, et al. Altered carnitine homeostasis is associated with decreased mitochondrial function and altered nitric oxide signaling in lambs with pulmonary hypertension Am J Physiol Lung Cell Mol Physiol, January 1, 2008; 294(1): L46 - L56. [Abstract] [Full Text] [PDF]

				H. Xu, Y. Shi, J. Wang, D. Jones, D. Weilrauch, R. Ying, B. Wakim, and K. A. Pritchard Jr. A Heat Shock Protein 90 Binding Domain in Endothelial Nitric-oxide Synthase Influences Enzyme Function J. Biol. Chem., December 28, 2007; 282(52): 37567 - 37574. [Abstract] [Full Text] [PDF]

				C. Blouet, F. Mariotti, V. Mathe, D. Tome, and J.-F. Huneau Nitric Oxide Bioavailability and Not Production Is First Altered During the Onset of Insulin Resistance in Sucrose-Fed Rats Experimental Biology and Medicine, December 1, 2007; 232(11): 1458 - 1464. [Abstract] [Full Text] [PDF]

				M. Duval, F. Le B uf, J. Huot, and J.-P. Gratton Src-mediated Phosphorylation of Hsp90 in Response to Vascular Endothelial Growth Factor (VEGF) Is Required for VEGF Receptor-2 Signaling to Endothelial NO Synthase Mol. Biol. Cell, November 1, 2007; 18(11): 4659 - 4668. [Abstract] [Full Text] [PDF]

				J.-L. Frossard, E. Schiffer, B. Cikirikcioglu, J. Bourquin, D. R. Morel, and C. M. Pastor Opposite regulation of endothelial NO synthase by HSP90 and caveolin in liver and lungs of rats with hepatopulmonary syndrome Am J Physiol Gastrointest Liver Physiol, October 1, 2007; 293(4): G864 - G870. [Abstract] [Full Text] [PDF]

				Y. Su, D. Kondrikov, and E. R. Block beta-Actin: A Regulator of NOS-3 Sci. Signal., September 18, 2007; 2007(404): pe52 - pe52. [Abstract] [Full Text] [PDF]

				B. Derakhshan, G. Hao, and S. S. Gross Balancing reactivity against selectivity: The evolution of protein S-nitrosylation as an effector of cell signaling by nitric oxide Cardiovasc Res, July 15, 2007; 75(2): 210 - 219. [Abstract] [Full Text] [PDF]

				D. M. Dudzinski and T. Michel Life history of eNOS: Partners and pathways Cardiovasc Res, July 15, 2007; 75(2): 247 - 260. [Abstract] [Full Text] [PDF]

				S. H. Oh, J. K. Woo, Y. D. Yazici, J. N. Myers, W.-Y. Kim, Q. Jin, S. S. Hong, H.-J. Park, Y.-G. Suh, K.-W. Kim, et al. Structural Basis for Depletion of Heat Shock Protein 90 Client Proteins by Deguelin J Natl Cancer Inst, June 20, 2007; 99(12): 949 - 961. [Abstract] [Full Text] [PDF]

				J. L. Aschner, S. L. Foster, M. Kaplowitz, Y. Zhang, H. Zeng, and C. D. Fike Heat shock protein 90 modulates endothelial nitric oxide synthase activity and vascular reactivity in the newborn piglet pulmonary circulation Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1515 - L1525. [Abstract] [Full Text] [PDF]

				F. Edlich, F. Erdmann, F. Jarczowski, M.-C. Moutty, M. Weiwad, and G. Fischer The Bcl-2 Regulator FKBP38-Calmodulin-Ca2+ Is Inhibited by Hsp90 J. Biol. Chem., May 25, 2007; 282(21): 15341 - 15348. [Abstract] [Full Text] [PDF]

				M. Garcia Blanes, M. Oubaha, Y. Rautureau, and J.-P. Gratton Phosphorylation of Tyrosine 801 of Vascular Endothelial Growth Factor Receptor-2 Is Necessary for Akt-dependent Endothelial Nitric-oxide Synthase Activation and Nitric Oxide Release from Endothelial Cells J. Biol. Chem., April 6, 2007; 282(14): 10660 - 10669. [Abstract] [Full Text] [PDF]

				P. Hawle, D. Horst, J. P. Bebelman, X. X. Yang, M. Siderius, and S. M. van der Vies Cdc37p Is Required for Stress-Induced High-Osmolarity Glycerol and Protein Kinase C Mitogen-Activated Protein Kinase Pathway Functionality by Interaction with Hog1p and Slt2p (Mpk1p) Eukaryot. Cell, March 1, 2007; 6(3): 521 - 532. [Abstract] [Full Text] [PDF]

				D. D. Roberts, J. S. Isenberg, L. A. Ridnour, and D. A. Wink Nitric Oxide and Its Gatekeeper Thrombospondin-1 in Tumor Angiogenesis Clin. Cancer Res., February 1, 2007; 13(3): 795 - 798. [Abstract] [Full Text] [PDF]

				M. Vercauteren, E. Remy, C. Devaux, B. Dautreaux, J.-P. Henry, F. Bauer, P. Mulder, R. Hooft van Huijsduijnen, A. Bombrun, C. Thuillez, et al. Improvement of Peripheral Endothelial Dysfunction by Protein Tyrosine Phosphatase Inhibitors in Heart Failure Circulation, December 5, 2006; 114(23): 2498 - 2507. [Abstract] [Full Text] [PDF]

				P. Hawle, M. Siepmann, A. Harst, M. Siderius, H. P. Reusch, and W. M. J. Obermann The Middle Domain of Hsp90 Acts as a Discriminator between Different Types of Client Proteins Mol. Cell. Biol., November 15, 2006; 26(22): 8385 - 8395. [Abstract] [Full Text] [PDF]